Figure 1.
Purification of p97/VCP and adaptor proteins.
His-tagged full length p97 or p97-N-D1 fragment or adaptor proteins Ufd1/Npl4 (UN) or p47 were expressed in E. coli Rosetta DE3 (Novagen) and purified under native conditions using nickel affinity chromatography and gel filtration. UN was co-purified as a heterodimer through the His-tag of Ufd1. (A) UV peak fractions at the appropriate elution volumes on Superose 6, corresponding to the respective oligomeric states of each protein that were used for Biacore binding assays (hexameric 13.5 ml fraction for full length p97/VCP and 14.5 ml fraction for p97-N-D1) (B) The purified proteins were resolved by SDS-PAGE and stained with Coomassie brilliant blue. For calibration of the Superose 6 column, see Figure S1.
Figure 2.
Binding affinities of the interactions between p97/VCP and its adaptor proteins.
p97/VCP was immobilized on a CM5 sensorchip surface, using amine coupling procedure, and a concentration series of p47 (two fold dilutions from 1.92 μM) or UN (two fold dilutions from 1.88 μM) were injected over the immobilized surface at a flow rate of 30 μl/min, at 25°C, using PBS as the sample and running buffer. Sensorgrams and the corresponding interaction maps for the p97/VCP-UN interaction without ATP (A), with ATP during UN association and dissociation (B) and for p97/VCP-p47 interactions when p47 is immobilized (C) and when p97/VCP is immobilized (D) are shown. Interaction Map Analysis shows that the interaction between UN and p97/VCP is complex, with a major component with ∼5 μM affinity, and a second component with ∼400 nM affinity. When ATP is added, the affinity of the second component increases to 100 nM. For the interaction between p97/VCP and p47, only one kinetic component was found, and the interaction could be fitted to a Langmuir 1∶1 interaction model, with ∼31.3 nM affinity when p47 is immobilized and p97 injected across (C) or ∼5 μM affinity when p97 is immobilized and p47 injected across (D).
Figure 3.
The design of the competition experiments using the Biacore 3000 instrument.
The cartoon depicts the different steps in each binding cycle designed to observe competition between Ufd1/Npl4 (UN) and p47 for binding to p97/VCP in real time. The first binding partner, p47, was immobilized on a CM5 sensorchip by amine-coupling procedure to a level of 2000 RU. Injection of p97/VCP across the p47 surface allowed the capture of p97/VCP to a level of 500 RU before UN was coinjected and p97/VCP dissociation was monitored. At the end of each experiment, the surface was regenerated by a 3 sec pulse of 50 mM NaOH. The capture of p97/VCP could then be repeated with UN coinjected at different concentrations or in the presence of ATP.
Figure 4.
Ufd1/Npl4 competes with p47 for binding to p97/VCP more effectively in the presence of ATP.
p47 (2000 RU) was amine-coupled onto a flowcell of a CM5 sensorchip and p97/VCP (500 RU) was repeatedly captured by the immobilized p47 as depicted in Figure 3 and described in Experimental Procedures. UN at the indicated concentrations was coinjected across the surface, in the absence (A) or presence (B) of 2 mM ATP. In control experiments, 2 mM ATP with no UN (C) or 5 µM BSA instead of UN (D) were coinjected. UN was coinjected at fixed concentrations of either 0.5 µM (E) or 3 µM (F) in the presence of the indicated ATP concentrations.
Figure 5.
ATP binding to the p97/VCP D1 domain regulates the adaptor protein competition.
Full length p97/VCP (A) or p97-N-D1 fragment (B) were captured on the p47 surface to a level of 500 RU. 5 μM of UN was coinjected across the surface in the presence of 1 mM of either ATP or ATPγS, using the same experimental setup described in Figure 4,
Figure 6.
ATP binding to p97/VCP affects Ufd1/Npl4 association with p97/VCP.
p97/VCP (1000 RU) was immobilized on a CM5 sensorchip and either UN (A)or p47 (B) were injected over the p97/VCP surface at a concentration of 0.15 μM in the absence or presence of 1 mM ATP during the binding and/or dissociation phases, as indicated. (C) 0.16 μM UN was injected across 1000 RU of immobilized p97/VCP in the absence or presence of 1 mM of ATP, AMP-PNP or ADP.
Figure 7.
Dynamic light scattering demonstates the stability of p97/VCP and p97-N-D1 hexameres.
Dynamic light scattering was performed on purified hexamers of either full length p97/VCP (A) or p97-N-D1 fragment (B) in the absence or presence of ATP. Upon addition of 1 mM ATP the hydrodynamic diameter of full length p97/VCP shifted from 19.41 nm to 19.74 nm and that of p97-N-D1 fragment shifted from 16.14 nm to 18.18 nm. The observed single peak homogeneities reflect no aggregation and stable hexamers. Both full length p97/VCP (C) and p97-N-D1 fragment (D) were subjected to thermal denaturation monitored by dynamic light scattering, showing no signs of hexamer dissociation when heated from 25°C to 70°C.
Figure 8.
Conformational changes upon ATP binding to p97/VCP and p97-N-D1 observed by differential scanning fluorimetry.
Hexamers of either full length p97/VCP (A) or p97-N-D1 fragment (B) were subjected to differential scanning fluorimetry in the absence or presence of 1 mM ATP. In the absence of ATP, the melting transition of full length p97/VCP hexamers was a two step process with the major component occurring at 60°C. Minor shifts were detected in the presence of ATP (A). p97-N-D1 hexamers had a melting temperature of 50°C in the absence of ATP, which was significantly shifted by 14°C to 64°C in the presence of ATP (B). Concentration-dependent thermostabilizations were observed by measurements of melting temperatures of full length p97/VCP (C) or p97-N-D1 fragment (D) at the indicated concentrations of ATP.